Can wrong disk geometry force a FAT32 "rebuild"?

[John Smith]

I have some questions about messing up a partiton which may have something
to do with disk geometry and Win98.

Using my PC (Duron-based) I copied 5 GB of data to a FAT32 partition on a
20GB PATA hard drive.

I took the hard drive and attached it as a slave to another PC (a 450 MHz
Compaq desktop) which runs Win98 from its own internal drive.

This second PC booted to Win98 oj its own drive but said the data on the new
drive's partition had some sort of file system error and it offered to
correct it. I said yes. The correction process went on for about 10
minutes before I stopped it. Messages were flashing by saying something
like such-and-such was in error and that it was correcting this. By the end
all the data was lost.

What is the likely cause of this?

The connection of 20GB PATA hard drive was correct because was later able to
save and read data to that same FAT32 partition using Win98 on the Compaq
without reformatting it.

I would have thought it likely that an error in the file system would cause
only a few things to be corrected. But this went on for ages. So I am
guessing that maybe a mismatched disk geometry cause the rejection of the
whole FAT32 partition.

Can someone advise please as I want to avoid this in future.

John

--

Sent to
comp.sys.ibm.pc.hardware.storage
microsoft.public.win98.gen_discussion
microsoft.public.win98.disks.general

 

[Tom Scales]

Did you ensure the jumpers are correct? On your Duron PC, it likely uses
Master/Slave. The Compaq likely uses Cable Select. If you left it Master
and put it on the Compaq, then it would have gotten royally confused.

Tom

 

[Rod Speed]

I have some questions about messing up a partiton which
may have something to do with disk geometry and Win98.

Using my PC (Duron-based) I copied 5 GB of data
to a FAT32 partition on a 20GB PATA hard drive.

I took the hard drive and attached it as a slave
to another PC (a 450 MHz Compaq desktop)
which runs Win98 from its own internal drive.

This second PC booted to Win98 oj its own drive but
said the data on the new drive's partition had some sort
of file system error and it offered to correct it. I said yes.

Not a good idea at all.

The correction process went on for about 10 minutes
before I stopped it. Messages were flashing by saying
something like such-and-such was in error and that it
was correcting this. By the end all the data was lost.

What is the likely cause of this?

Most likely the drive type entry in the bios was
different on the original system and the compaq.

The connection of 20GB PATA hard drive was
correct because was later able to save and
read data to that same FAT32 partition using
Win98 on the Compaq without reformatting it.

I would have thought it likely that an error in the file
system would cause only a few things to be corrected.

Not when the drive type entry is different on the two systems.
That affects the mapping of CHS values to LBA values and
you can get one hell of a mess with some remappings.

But this went on for ages. So I am guessing
that maybe a mismatched disk geometry cause
the rejection of the whole FAT32 partition.

Or it was a considerable mess in the Compaq and
the purported 'repair' made that mess much worse.

Can someone advise please as I want to avoid this in future.

Just ensure that you are using an AUTO drive type in both systems.

And when there is an initial complaint, DONT allow it to repair
and dont write to the partition at all until the problem is resolved.

 

[Folkert Rienstra]

Did you ensure the jumpers are correct? On your Duron PC, it likely uses
Master/Slave. The Compaq likely uses Cable Select. If you left it Master
and put it on the Compaq, then it would have gotten royally confused.

Nonsense. The setting is on the drive, not the PC.

 

[Folkert Rienstra]

Not a good idea at all.



Most likely the drive type entry in the bios was
different on the original system and the compaq.

That is only used for booting.
It doesn't matter for nonboot(able)(ed) drives.

Not when the drive type entry is different on the two systems.

If it mattered you wouldn't even see volume structures.

That affects the mapping of CHS values to LBA values and
you can get one hell of a mess with some remappings.

Only the LBA settings in the MBR are interpreted.

Or it was a considerable mess in the Compaq and
the purported 'repair' made that mess much worse.

Much more likely.

 

[Rod Speed]

Nonsense. The setting is on the drive, not the PC.

Wrong. As always.

 

[Rod Speed]

That is only used for booting.
It doesn't matter for nonboot(able)(ed) drives.

Wrong. As always.

If it mattered you wouldn't even see volume structures.

Wrong. As always.

Only the LBA settings in the MBR are interpreted.

Wrong. As always.

Much more likely.

Doesnt explain why problems were seen that needed repairing.

 

[Steve Baron - KB3MM]

Yes but the cables are in the PC!

Nonsense. The setting is on the drive, not the PC.

 

[AlmostBob]

cable select/master/slave refers ONLY to the jumper settings on the back of
the drive, the settings really should match so if you are plugging in the
master connector, set the jumper to master, slave, slave jumper, unsure set
jumper to cable select. here is a picture or two one shows the location of the
jumper on the end of the drive,
http://www.coloredhome.com/PC/PCassembly51051.JPG the other a close up of the
jumper
http://www.coloredhome.com/PC/PCassembly510511.JPG

--
Adaware http://www.lavasoft.de
spybot http://security.kolla.de
AVG http://www.grisoft.com
Panda online scan http://www.pandasoftware.com/ActiveScan/
Catalog of removal tools http://www.pandasoftware.com/download/utilities/
Blocking Unwanted Parasites with a Hosts file
http://mvps.org/winhelp2002/hosts.htm
links provided as a courtesy,
Grateful thanks to the authors/webmasters


| Yes but the cables are in the PC!
|
| | | > > Did you ensure the jumpers are correct? On your Duron PC, it likely
| uses
| > > Master/Slave. The Compaq likely uses Cable Select. If you left it
| Master
| > > and put it on the Compaq, then it would have gotten royally confused.
| >
| > Nonsense. The setting is on the drive, not the PC.
| >
| > >
| > > Tom
| | > > > I have some questions about messing up a partiton which may have
| something
| > > > to do with disk geometry and Win98.
| > > >
| > > > Using my PC (Duron-based) I copied 5 GB of data to a FAT32 partition
| on a
| > > > 20GB PATA hard drive.
| > > >
| > > > I took the hard drive and attached it as a slave to another PC (a 450
| MHz
| > > > Compaq desktop) which runs Win98 from its own internal drive.
| > > >
| > > > This second PC booted to Win98 oj its own drive but said the data on
| the new
| > > > drive's partition had some sort of file system error and it offered to
| > > > correct it. I said yes. The correction process went on for about 10
| > > > minutes before I stopped it. Messages were flashing by saying
| something
| > > > like such-and-such was in error and that it was correcting this. By
| the end
| > > > all the data was lost.
| > > >
| > > > What is the likely cause of this?
| > > >
| > > > The connection of 20GB PATA hard drive was correct because was later
| able to
| > > > save and read data to that same FAT32 partition using Win98 on the
| Compaq
| > > > without reformatting it.
| > > >
| > > > I would have thought it likely that an error in the file system would
| cause
| > > > only a few things to be corrected. But this went on for ages. So I
| am
| > > > guessing that maybe a mismatched disk geometry cause the rejection of
| the
| > > > whole FAT32 partition.
| > > >
| > > > Can someone advise please as I want to avoid this in future.
| > > >
| > > > John
| > > >
| > > > --
| > > >
| > > > Sent to
| > > > comp.sys.ibm.pc.hardware.storage
| > > > microsoft.public.win98.gen_discussion
| > > > microsoft.public.win98.disks.general
|

 

[Rod Speed]

cable select/master/slave refers ONLY to
the jumper settings on the back of the drive,

Wrong. The correct cable is needed for cable select too.

And the bios needs to be able to handle both approaches too.

the settings really should match so if you are plugging
in the master connector, set the jumper to master,
slave, slave jumper, unsure set jumper to cable select.

Wont work if it aint a cable select cable.

here is a picture or two one shows the location
of the jumper on the end of the drive,
http://www.coloredhome.com/PC/PCassembly51051.JPG
the other a close up of the jumper
http://www.coloredhome.com/PC/PCassembly510511.JPG

Irrelevant to what is being discussed.

 

[Folkert Rienstra]

Yes but the cables are in the PC!

Who said anything about cables? Now re-read that sentence.

[snip]

 

[Folkert Rienstra]

Wrong. The correct cable is needed for cable select too.
Bullshit.


And the bios needs to be able to handle both approaches too.

Absolutely clueless, "as always".

That is nonsense.
There are no master and slave connectors on any cable except for cableselect
when the cable position *determines* what will become master and what slave.
Master and slave *jumpered* drives can be on either connector.
With only one drive put it at the end connector with any setting that you like,
M, S or CS.

unsure set jumper to cable select.

Unsure means that you must make an effort to *become* sure and
not set the drive to cableselect unless you set them both that way.
Other wise you may end-up with 2 masters or two slaves.
(Again with single drives it doesn't matter what you choose).

Wont work if it aint a cable select cable.

Works fine with one drive per cable.
Works fine too with two drives if the other drive is set to slave.

Irrelevant to what is being discussed.

[snip]

 

[Rod Speed]

Some pathetic excuse for a troll claiming to be
just the pig ignorant trolling thats all it can ever manage.

 

[Steve Baron - KB3MM]

The master slave setting is not on the drive for cable select

Yes but the cables are in the PC!

Who said anything about cables? Now re-read that sentence.
something

[snip]

 

[Folkert Rienstra]

The master slave setting is not on the drive for cable select

You don't say. Maybe that is why they call it "cable select" then, eh.

What exactly did you not understand in:

" If you left it Master and put it on the Compaq, then it would have gotten royally confused".

Answer: It will be 'Master' on the Compaq, regardles of *any* cable.
No confusion *at all*.

Nonsense. The setting is on the drive, not the PC.

Yes but the cables are in the PC!

Who said anything about cables? Now re-read that sentence.

Did you ensure the jumpers are correct? On your Duron PC, it likely uses
Master/Slave. The Compaq likely uses Cable Select. If you left it Master
and put it on the Compaq, then it would have gotten royally confused.

Nonsense. The setting is on the drive, not the PC.


Tom
I have some questions about messing up a partiton which may have something

[snip]

 

[Steve Baron - KB3MM]

WOW !

Holey moley !

The master slave setting is not on the drive for cable select

You don't say. Maybe that is why they call it "cable select" then, eh.

What exactly did you not understand in:

" If you left it Master and put it on the Compaq, then it would have gotten royally confused".

Answer: It will be 'Master' on the Compaq, regardles of *any* cable.
No confusion *at all*.

Nonsense. The setting is on the drive, not the PC.

Yes but the cables are in the PC!

Who said anything about cables? Now re-read that sentence.


Did you ensure the jumpers are correct? On your Duron PC, it likely uses
Master/Slave. The Compaq likely uses Cable Select. If you left it Master
and put it on the Compaq, then it would have gotten royally confused.

Nonsense. The setting is on the drive, not the PC.


Tom
I have some questions about messing up a partiton which may have something

[snip]

 

[Tom Scales]

My statement, poorly and incorrectly quoted:

If you left it Master and put it on the Compaq, then it would have

gotten royally confused

was NOT because it was a Compaq, but because COMPAQ's use Cable Select
Cables.

Tom

WOW !

Holey moley !

You don't say. Maybe that is why they call it "cable select" then, eh.

What exactly did you not understand in:

" If you left it Master and put it on the Compaq, then it would have gotten royally confused".

Answer: It will be 'Master' on the Compaq, regardles of *any* cable.
No confusion *at all*.

left
it Master

and put it on the Compaq, then it would have gotten royally confused.

Nonsense. The setting is on the drive, not the PC.


Tom
I have some questions about messing up a partiton which may have something

[snip]

 

[Folkert Rienstra]

My statement, poorly and incorrectly quoted:

Yeah, that bloke obviously has no idea how to set-up his newsreader properly.
Oh btw, since you mentioned poor quoting, any idea why the word 'left' is on
a new line -all by itself- (line 51) in your post?

If you left it Master and put it on the Compaq, then it would have

Ah yeah, you quoted that *soooo* much better.

was NOT because it was a Compaq, but because COMPAQ's use Cable Select
Cables.

So does *every other computer that uses ATA66 and higher*.
By your elaborate thinking it must have been already "royally confused" in the
original computer, yeah?

Tom

WOW !

Holey moley !

The master slave setting is not on the drive for cable select

You don't say. Maybe that is why they call it "cable select" then, eh.

What exactly did you not understand in:

" If you left it Master and put it on the Compaq, then it would have gotten royally confused".

Answer: It will be 'Master' on the Compaq, regardles of *any* cable.
No confusion *at all*.


Nonsense. The setting is on the drive, not the PC.

Yes but the cables are in the PC!

Who said anything about cables? Now re-read that sentence.


Did you ensure the jumpers are correct? On your Duron PC, it likely uses
Master/Slave. The Compaq likely uses Cable Select. If you left it Master
and put it on the Compaq, then it would have gotten royally confused.

Nonsense. The setting is on the drive, not the PC.


Tom
I have some questions about messing up a partiton which may have something

[snip]